Inheritance of the apolipoprotein (apo) E4 allele is the major identified genetic risk factor for sporadic Alzheimer's disease (AD). We suggested in 1992 that apoE, especially its apoE4 isoform, can promote the conformational change of amyloid beta (Abeta) from soluble Abeta to toxic aggregates and amyloid plaques and coined the term "pathological chaperone" to describe this role. Many subsequent studies have been consistent with this hypothesis. In this proposal, based on our preliminary data, we plan to develop a novel, non-toxic therapeutic strategy for amyloid clearance in Alzheimer's disease, by blocking the apoE/Aa interaction. Based on the known binding site of apoE on Aa (Aa residues 12-28), we synthesized an end-protected, D-amino acid peptide Abeta12-28P, where the valine at residue 18 is substituted by proline. This modification renders the peptide non-toxic and non-fibrillogenic, without affecting binding to apoE. Abeta12-28P is blood-brain barrier permeable (clearance=65q20ul serum/g) and its serum half-life is 62q18 min. Our data shows this peptide reduces the Abeta/apoE toxicity and fibril formation in vitro. Transgenic APP/PS1 AD mice treated for one month with Abeta12-28P have a 63% reduced amyloid burden in the cortex (p=0.0048) and 62% in the hippocampus (p=0.0047) compared to controls. No toxicity was observed in the animals and no humoral response to Abeta was detected under our experimental conditions (so the treatment effect could not be due to antibody clearance of Abeta). Our preliminary data also indicates this treatment can reduce or remove existing amyloid lesions as documented by in vivo 2-photon imaging. Our Specific Aims are: 1) To evaluate the effect of blocking the apoE/Abeta interaction on amyloid burden (by in vivo imaging and histology), neuronal pathology and behavior in Tg animal models of Alzheimer's disease. 2) To identify the Aa binding site on apo E, so that further inhibitors of the apoE/Aa interaction can be designed based on the apoE sequence where Abeta binds and to develop peptidomimetics based on both the Aa and apoE sequences. 3) To determine the effects of blocking the Abeta/apoE interaction on other apoE functions as assessed by hippocampal neuronal spine integrity and receptor binding. These studies will test our central hypothesis that blocking the Aa/apoE interaction can serve as a novel, non-toxic therapeutic target for Alzheimer's disease.